ResearchDatasetsHotpotQA, RULER-QA, NIAH, 2Wikimqa, Musique, Narrativeqa, QasperFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsLong-context UnderstandingHotpotQA, RULER-QA, NIAH, 2Wikimqa, Musique, Narrativeqa, Qasper (test)47.66HotpotQA Score16
Long-context UnderstandingHotpotQA, RULER-QA, NIAH, 2Wikimqa, Musique, Narrativeqa, Qasper (test)47.66HotpotQA Score16